Urtext = Plain Text ? – An Analysis of the Sarabande in Eb Major, Part 1 (Bar 1-12)


This is a translation of the post

Urtext = Klartext? – eine Analyse der Sarabande in Es-Dur, Teil 1 (Takt 1-12)

by Dr. Marshall Tuttle.

Michael Bach

This is the first part of the analysis of the

Sarabande in E Flat Major

with performance examples

and the interpretations of the cello competition *)



Interpretation of the “Sarabande in Es-Dur”
Michael Bach, Cello with BACH.Bow




Sarabande Es-Dur AMB-
Copy by Anna Magdalena Bach, digital scan of the Staatsbibliothek in Berlin – PK



JSB = Johann Sebastian Bach
AMB = Anna Magdalena Bach

2-slur, 3-slur, 4-slur = Slurs over 2, 3 or 4 notes.
Three 2-slurs = three slurs over 2 notes
m, mm = measure, measures

C4 = middle C

Sub = Subdominant
Dom = Dominant
Rel = Relative minor
SR = Relative minor of the subdominant
DR = Relative minor of the Dominant
DD = Dominant of the Dominant
D7-chord= Dominant seventh chord
D9-chord = Dominant ninth chord
s6-chord = Minor subdominant with sixth instead of fifth (ii6 chord)
SD = Secondary dominant
SDD = Dominant of the secondary Dominant


The analysis of the “Sarabande in E Flat Major” by JSB presented here is offered in comparison to the individual interpretations of the participants in this year’s Cello Competition for New Music in Stuttgart* ). This is to elucidate the extent to which today’s current interpretations do justice to the original manuscript. In this “Sarabande” there are no significant deviations in the Urtext editions from AMB’s copy, and there are no scribal errors on the part of AMB. Nevertheless, a performance convention has been established that modifies the musical text. This relates to the polyphony, to the slurs and to the addition of “ornaments”.

It does not seem to be common knowledge (even the Urtext editions do not consider the following) that in polyphonic passages the longer note values dictate the bow stroke. That is, the sounds of other voices which have shorter note values, are slurred accordingly, for example, here in the opening bars.

More important than the slurs, however, are the accents that are implicitly notated. Consequently, the slurs allow clear conclusions to agogical and dynamic processes. Thus, JSB therein had an effective, and at the same time flexible notational symbol at hand.

In general, the slurs are identical to the bow strokes. Nevertheless, for technical reasons, a stroke of the bow can be changed once. The key remains that the meaning of a “compositional” slur is recognized and the resultant sound is implemented accordingly. Nevertheless, the slurs are in the “Suites” almost always identical to the optimal bow strokes. The two exceptions are in this “Sarabande” measures 29f

Ties do not cause accentuation, as they merely extend a pitch. However, in this movement something else is very striking, namely that JSB decidedly notated ties across the barline. In some interpretations this resulted in the misconception that a tie is missing in the first and third measures on the pitch Ab3 and Db4, which would have to be supplemented as well as sometimes in similar looking places.

And so, once again we are brought back to the harmony. Because only deep harmonic analysis gives us the crucial insight as to why no ties are present, for example, in the first 4 measures. The formative ties in this movement are therefore to be understood less as articulatory binding, but rather as a sign that creates harmonic meaning. In the cello competition a mistaken understanding of the meaning of the tie, resulted in ties often being applied to all similar looking musical motives without consideration of their harmonic context.

The usage of ties in this “Sarabande” also explains, for example, why in m 23 the s6-chord is articulated rather than the DD7-chord. This makes it possible to tie the note Eb3 to the note Eb3 of the first beat of the following measure, and has an impact on the weighting of the individual chord tones with one another. The “absence” of a tie in the opening measures, for example, gives us the indication that the seventh of a Dom chord (Ab3 in m 3), and the suspended fourth of its resolution (in musical notation the same note Ab3) is, strictly speaking, not the same pitch frequency. This leads to tangential topics such as micro-tonality.which deal with issues beyond the intonation of tempered tuning.

The key of E flat major is strengthened in the opening measures of the “Sarabande” with its Sub and Dom But the DD, expected in m 8, is “denied” in the first half of the movement, a fact which disturbs the listener. The DD occurs later in m 21, “belatedly” so to speak, and in conjunction with the opening motif, but only once, then all the more surprising and heroic.

In contrast, the minor keys play a dominant role, although they are more likely to remain in the background. They are often confused with the Dom or the DD. They (the minor keys) bring something like anexus of harmonic gravity, but their chords do not explicitly appear.

An outstanding feature is the trill at the end of the first half of the movement. It can be demonstrated that in the Cello Suites no ornaments exist that can be understood in the traditional sense as beautifying or decorative embellishments. Trills in the Cello Suites are always an indication of two or more oscillating harmonies. This trill on the root of the Dom, not on its third, also creates an open end to the first half of the movement, because its resolution is missing. The somewhat clueless trilling of some of the artists in the cello competition is discussed below.

The Curved-Bow-specific issue is not presented in this analysis, although the included sound examples are performed by me with the curved bow. However, there are two places that are definitely not feasible without the curved bow: mm 30f. For other passages the curved bow offers various modes of bowing that correspond to the compositional structure according to the motto: the best instrument is just good enough for JSB.

The polyphony, especially with regard to the note values, is precisely notated. JSB has written nothing alterable in these works. Every pitch and every note value, as well as every slur and every trill has its explanation. The transcript of AMB bears witness.


The Slur-Code

Here I codify some insights that have emerged from my detailed analyses of all the “suites” in the copy of AMB, almost as axioms advance. I call them the “slur-code” in the works for solo cello. In this Sarabande, the following rules apply:


The first note under a slur is emphasized, as is the first note after the slur. The longer the slur, the stronger the subsequent noteis emphasized and the weaker is the first note of the slur. This sometimes leads to the extreme case that an exceptionally long slur has to start very quietly, while the subsequent pitch is very strongly accentuated (e.g. m 13 of the “Allemande” in G Major).


When one slur directly follows another, the first note of the second slur receives emphasis, and the subsequent, linked notes decrescendo. The first note after the second slur is consequently unstressed.


Slurs are not notated when the polyphony makes their necessity clear.


Analysis of the “Sarabande in E Flat Major
… and its interpretations in the Cello Competition *)


Sarabande Es Edition MB 001

Sarabande in Eb Major“, Edition Michael Bach
Sheet music example mm 1-4

Slurs that are understood from the polyphony and which JSB therefore did not have to notate are added in parentheses.


M 1

The open fifth at the beginning lacks the third. (This harmonic ambiguity is in contrast to m 21, where the same motive ipresented in DD with its third). Only with the second note, C4, is the major key clear. The following seventh Db4, sounding with the bass note Eb3, discloses the dominant character of this initial measure. The upper voice is slurred, because in the lower voice a dotted half note is notated.

M 2

The first beat is emphasized due to the three-part chord and, according to the “Slur Code”, because the relatively long slur of of the previous measure naturally creates an emphasis. Also in the first beat of m 2, the upper voice is slurred, since a quarter is notated in the lower voices. According to the “Slur Code”, this first beat gets a decrescendo. The suspension resolves to the third C4 in the second beat, which is thus unstressed.

M 3

This measure begins with an upbeat of three slurred sixteenth notes. Thus, the first beat of m 3 is at this time, in contrast to m 1, emphasized. It follows that the entire m 3 receives a decrescendo.

M 4

It follows further that the first beat of m 4 is now unstressed, despite the three-part chord. The resolution of the suspension to G3 on the second beat is now anticipated after the first two measures by the listener. On account of the slur in the first beat, the resolution to the third G2 of the tonic on the second beat is then slightly accented. The tonic is for the first time clearly established.


In m 2, the short decrescendo and crescendo are balanced so that the m 3 returns to its end at about the samevolume as the beginning of the phrase. Had JSB intended to repeat the dynamics (ie were a crescendo in m 3 intended) the three sixteenth notes would not be slurred. But a repeat of the dynamics of mm 1 and 2 would contradict the harmonic development tonic-Sub-Dom-tonic, which relaxes with the resolution to the tonic in m 4. In addition, a repeated crescendo and decrescendo in these 4 measures would structure the phrase into two 2-measure units that would split the cadential introduction. Particularly revealing, therefore, is the emphasis on each of the first beats of mm 2 and 3 (Sub and Dom), the unstressed first beats of mm 1 and 4, and the accent on the third of the tonic in m 4, so that the cadential conclusion is precisely located.

To further highlight the emphasis on the Dom, a chord in m 3 would be conceivable but illogical, since the subject is marked by the sound of an open fifth without a third. A chord with third at the beginning of this motif is reserved solely for the DD in m 21. A chord in m 4, however, is inevitable on the cello, as a result of the leap of a fifth downward in the bass, because the middle string sounds inevitably if the two outer strings are played. The bass note Eb2 is necessary to specify the suspended fourth in the upper voice.

These four opening bars show precisely how JSB intended the design of the opening cadence. He just needed to write down a single slur, on the three sixteenth notes of m 2. The other slurs, including accents and dynamics, arise naturally from the notated polyphony.

The opening cadence can be seen as a harmonic unit. It is the most prominent feature of this movement that is persistently avoided, except for a few prominent places, new beginnings or separations. Not even the two halves of the movement are clearly separated., because tthe trill destroys a clear resolution to the Dom and does not indicate a stablefinal note.

Moreover: For the repeated notes in the upper voice (the seventh Db4 and the suspended fourth Db4, and the seventh and fourth Ab3 respectively) each is strictly speaking not the same pitch. You could even intone the Seventh slightly lower than the fourth, as a kind of overtone of the fundamental, because it sounds along with the root. (In just intonation the pitch relationship of the two notes Ab3 in the mm 3 and 4 would be the fourteenth (seventh plus fifth) 3/2 x 7/4 = 21/8 = 63/24, and the eleventh (octave plus quart) 8/3 = 64/24. At a frequency of the root note of 78 Hz = Eb2, the Dom of the seventh would be 204.75 Hz and the fourth of the octave would be 208 Hz, which means a difference of about a quarter-tone. (In this pitch range the semitone is approx. 12 Hz.



How was this phrase performed in the cello competition?

The performances in the competition rarely implemented the two voices and the legato in the upper voice in mm 1 and 3. Never was the upper voice slurred in the first beats of mm 2 and 4.Sometimes the seventh on the third beat was mistakenly tied to the suspended fourth in the following measure simply by omission of the repetition of the tone in the upper voice on the first beat. Thus, the syncopated patterns of other places was approximated.(see Preface), because with the concave bow only a fictional slurring can be imitated. Bowing the two bass notes requires the upper note to be abandoned.

In addition, all participants in the cello competition demonstrated variants leading to shortening of the bass notes that would otherwise be repeated due to a bow change. But, the shortening of the bass notes in the performances is neither justified by an analysis of the score nor by any performance traditions. (It was also not taken into account that in JSB ‘s notation, mm 13 and 21 do contain shorter bass notes, in contrast to mm 1 and 3. There is a differentiation whose harmonic sense will be explained at the appropriate time.)



Sarabande Es Edition MB 002
Sheet music example mm 5-8


The next passage up to and including m 7 now shows why twice a tie occurs over the bar line, but not in m 6. The note Eb3 in m 4 is connected to the first beat of the following measure, because the tonic remains unchanged. A change of harmony or resolution does not happen at this time. In the lower part the 1/8-note Bb2 appears as fifth of the tonic chord. You can not interpret this note as the root of the Dom because the third is missing, and JSB shortened the Bb2 to an 1/8-Note. Therefore the note C3 in the upper voice of the first beat is not tied, and so the note D3 on the second beat is not accented. The dominant function remains for the time being in limbo. Only with the note Ab3 on the third beat, similar to the first and third measures, is the dominant audibly established.

It is perhaps perhaps somewhat speculative at this early point in the “Sarabande“, but a faint foreshadowing of the D Rel might already be intended with the notes Bb2 and D3. The D Rel is never achieved in the entire movement, but it is latently present in many places. Instead of D Rel to suspect the T Rel here would be even slightly more likely, because the sixteenth note C3 is not connected and, together with the Eb3, can be heard as a chord tone of T Rel. These considerations mean that with any new entering pitch of m 5 different harmony can be expected: with C3 the T Rel; with D3,the D Rel; with F3 the Dom Only the tritone D3-Ab3 clarifies the harmonic situation finally to the Dom With the syncopated tie the note Eb3 thus creating a “harmonic vacuum” that occurs directly after the opening cadence. This is puzzling, and is a powerful dramatic agent that is developed more fully in m 24.

M 6

Measure 6 begins with two harmonies, the tonic (Eb2 -Bb2) and the Dom (Bb2- Ab3) shown by articulation and the note values of the chord. The third beat of the preceding measure, here the seventh Ab3, is not tied into the next measure as it was inm 4 to m 5 previously. The articulation does not follow the motivic patterns of mm 2 and 4 and does not lead to the familiar harmonic sequence, a resolution on the second beat. The subtle difference in notation can be found in the shortening of the two lower notes to a dotted eighth, making the remaining sixteenth note F3 a chord tone, the fifth of the Dom Thus again, the second beat (note G3) is unstressed, so that a resolution to the tonic will be placed on hold.

In order to express the harmonic ambivalence of this chord, it is advantageous to play all three notes of the chord simultaneously and of equal duration, as an unresolved dissonance.

With the second beat (note G3) the tonic is apparently achieved, but the top note Db4 on the 3rd beat acts as seventh. The tonic is transformed into a SD. But,due to the fact that there is no Eb3 notated in the bass, JSB has not yet clarified a final unambiguous position on harmonic function of this chord.

M 7

The seventh Db4 is now tied to m 7. In the lower voice surprisingly, the note E3 appears. At first, it is unclear which of the two notes is dissonant. Theoretically, the note E3 could be a neighbor tone, which would return to Eb3. And so, the note Db4 would still be the seventh of the previous measure which could end up in a resolution to the Sub, as in m 2.

But the lower voice does not resolve, and the further course shows with confirming repetition of the third E4 that the harmony of the measure has been transformed, with the notes Db4 and E3 to a D9-chord, namely the SD of the DD. The note E3 is the third. It is indeed shortened to a dotted eighth note but should be fully sustained in order to highlight this surprising twist.


The tieing of the note Db4 actually suggests the continuation of the existing harmony. Nevertheless, a change of harmony takes place. What is interesting is that the seventh of the tonic Db4 (m 6) and the Ninth Db4 of the SD of the DD (in m 7 ) have the same frequency and their slurring is thus possible. (The interval ratio of the fundamentals of the SD of the DD (C3) and of the tonic (Eb4) is a minor third, ie 6/5, and therefore in just intonation the frequency of this natural seventh calculated from C3 would be 6/5 x 7/4 = 21 /10 and that of the ninth of the SD of the DD (fifth and tritone) would be 3/2 x 7/5 = 21/10. Without the use of the 7th partiall calculated because the seventh does not sound together with the root note: minor seventh over minor third 6/5 x 16/9 = 32/15 and octave plus minor second 2 x 16/15 = 32 /15.

Thus, the meaning of harmonic ties over the barline is comprehensible. It becomes clear how closelythe requirements of just intonationthrough pitch relations on integer ratios are related to harmonic processes. JSB was clearly aware of this.

The question of why the lower voice in m 5 is reduced to an eighth note while in mm 6 and 7 only to a dotted eighth note is explained by the fact that in m 5 the harmony is consonant and a harmonic change is only indicated but does not take place, whereas the chords in the first beats of mm 6 and 7 are dissonant. But finally,they leave nothing to the direction in which they are resolved. With the shortening of the bass notes by a sixteenth note an accentuation of the second beat is also avoided. The emphasis in mm 5 to 7 is instead rather on the 3rd beat, including the upbeat sixteenth note, because the new harmony only manifests itself at this point.


M 8

After an ostentatious break due to a sudden octave leap down, the answer to this highly demanding harmonic development remains in the future. Actually, one expects the final DD. However, what sounds is the dissonant fourth F3-Bb3, and it is again unclear which tone will “resolve”. In theory the transformation of the lower note, so here the note F3 to the E3 of the previous measure is conceivable. This would mean a further sounding of the SD.

But there follows the resolution of Bb3 to A4 so that the DD would be much more realistic. However, an extremely unusual slur is now on these two notes, so that the resolution on the 2nd beat is not being highlighted. It is one of the very rare cases in the “Suites”, where the resolution is attached to a suspension. Also, the lower note F3 is not continued in the 2nd beat, which would consolidate the DD, and which would be easily playable.

Both the slur and the “thrifty” quarter bass note F3 cause the slurred note A3 to appear not as a resolution but as a passing tone. The DD becomes “quite small”. Another possible goal would be a false resolution to the minor of the DD, i.e. D minor. This also suggests that the “Sarabande” develops further on in the minor keys.

In any case, JSB frustrates a triumphant conclusion after the three-bar increase and the declamatory turning point. The last sixteenth note Eb3 evaporates over into the next cycle.

Note:This harmonic process is very reminiscent of the “undermining” of the Rel in m 14 of the “Prelude” in G Major (see analysis of the ” Prelude“, part 2).



How was this phrase performed in the cello competition?


Some performers shortened the note values in the lower voice, some elongated the note F3 in m 8, to illustrate the resolution to the DD. Some eliminated even the notated Ab3 in the upper part of the 1st beat in m 6 to simulate a syncopated tie because the concave bow does not allow three voices here. The dynamic development here, i.e. the 3-bar increase and the “scared” withdrawal in m 8 remained unclear, and so the distinct caesura before m 8 was rarely realized.


Sarabande Es Edition MB 003
Sheet music example mm 9-12

M 9

As described, the “resolution” to DD in m 8 of the composition is not presented very convincingly. The fluctuation or blurring of harmonies is now continued by JSB. In the last passage of the first half the DD, the Dom, and the tonic, establish a close relationship. Their relative minors, however, occur in subsequent measures in an equally harmonious action. Measure 9 is probably in the Dom, although a touch of Rel of the DD is maintained from the measure before, because of the D3 sounding bass note. In this interpretation, the note Bb3 in the upper voice would be an added sixth, which in the second beat resolves to A3 (sixteenth note). The quarter note D3 in the bass implies that the octave D4 in the upper part is slurred and the top note F4 is to be emphasized.

M 10

The lower voice Eb3 in conjunction with the upper voice Bb3 in m 10 suggests the tonic. The fact that it also has a quarter note value, causes the upper D4 to be heard as a dissonance (major seventh). The top note G4 is particularly emphasized as a result of the longer tied two beats.

This dissonance Eb3-D4 shows the harmonic ambivalence of this measure, because it could also be interpreted as Dom, with a note Eb3 as a passing tone in the bass. Or, what moves during the “Sarabande” more and more in the realm of possibility, the DR appears here. The note Eb3 would be an added sixth in this interpretation. (In the second half of the movement, the Rel is obvious in m 26 because this harmony has already been clearly introduced in m 15.)

Incidentally, this dissonance Eb-D4 in the first beat is taken up again and continued as a tonal characteristic in m 27ff in the second half of the movement. Anyway, it should sound clearly, because the following high note G4 shines particularly radiantly.

M 11

The Eb3 from the previous measure is not resolved in this measure to D3 but continues to the note F3. Since the lower part is a quarter note, the upper voice is tied. Now the note A3 sounds together with F3 (as opposed to m 8). The note A3 belongs to the Dom and the DD, but also to their Rel. The following emphasized F3 on the 2nd beat can extent the implication of the DD or the Rel of the DD.



From m 8 there are ambiguous harmonic interpretations conceivable which are parallel. Simplifying is a possible progression in major: DD – Dom – tonic – (DD) – D (tr ?). The other possibility is in the minor: Rel of the DD – Rel of the D – Rel of the DD -Rel of the D (tr). This major-minor ambiguity dominates the entire second half of the movement, leading to the harmonically exuberant five closing bars.


M 12

The trill C3-Bb2 on the final note of the first half of the movement obscures the arrival of the Dom, or D Rel. Besides, how is it actually to be understood: Is this a trill on the root of the Dom? Can it be understood as a trill on the third Bb2 of D Rel, or even as an indication of C3, the root of Rel. This trill acts “unreal” because it follows no regular resolution. The note that would provide completion or “closing harmony” is obviously lacking. Perhaps we can explain this trill only in retrospect, after analysis of the second half of the movement. For it will become more and more obvious, such as the mm 28 and 32, that the chords of two latent harmonies often differ in only two pitches. This leads to the question, to what pitch should this Bb2 trill, C3 or A2? The first option sounds like “Major”, the second like “minor”, which would further lead to the D Rel. But more likely is the first option, because,analogous to the last measure, it turns back to the major keys. In this case, the note C3 would be unique to the DD, the note Bb2 would then stand for the Dom So the trill of m 12 represents a “flicker” of the DD, which still maintains its presence.




How was this phrase performed in the cello competition?

All dotted rhythms were, uniformly, played exactly as in the previous 4 measures, ie the lower voice was reduced to an eighth note and the upper voice was not slurred.

Thus, the intervals D3-D4, Eb3-D4 and F3-A3 were not performed, with an expressive legato in the upper voice. The slurs on the respective first beats were not necessary for JSB to notate, as mentioned, due to the quarter notes in the second voice. In m 11 no emphasis was detected on the note F3 on the 2nd beat. With two exceptions, the trill was ended on the principal note, a more or less sustained B2, which is not notated.



Cello-Competition for New Music and award of the Domnick Cello Prize

Staatliche Hochschule für Musik und Darstellende Kunst Stuttgart, Germany

January 20-25, 2014


The Participants:

Charles-Antoine Duflot
Raphael Moraly
Armand Fauchère
Magdalena Bojanowicz
Darima Tcyrempilova
David Eggert
Hugo Rannou
Minji Won
Beatrice Holzer-Graf
Hanna Magdaleine Kölbel
Jee Hye Bae






Leave a Reply